From pharmacokinetic studies in animals and man and in vitro binding studies we have shown that acyl glucuronide metabolites of bilirubin and zomepirac (ZomaxR) react nonenzymicly with serum albumin to give covalent adducts. For bilirubin, substantial quantities of adducts are formed only when normal excretion of bilirubin glucuronides in bile is impaired as in patients with cholestatic jaundice. With zomepirac, however, covalent binding to serum proteins via the acyl glucuronide occurs even in healthy humans after a single drug dose. If plasma clearance of zomepirac and its glucuronide is reduced in humans by concurrent probenecid administration, or in animals by experimentally-induced cholestasis, formation of zomepirac-albumin adducts increases. These observations coupled with the known chemical reactivity of acyl glucuronides and the common occurrence of such compounds in vivo as drug metabolites, suggest that covalent binding of carboxylic acids to albumin, or to other proteins and macromolecules, via glucuronide metabolites may be a general phenomenon of wide significance. We propose to test this hypothesis, which if true could have major pharmacological and toxicological significance, and to determine whether factors, such as age, disease states, chronic dosing and the stability of the glucuronide, influence the in vivo exposure of animals or humans to these reactive acyl glucuronides. Purified acyl glucuronides from several acidic drugs (eg. salicylate, ibuprofen, diflunisal, furosemide, tolmetin, zomepirac and benoxaprofen) will be incubated with serum albumin in vitro and the extent of covalent binding will be measured. Zomepirac glucuronide and other reactive acyl glucuronides will be examined for similar binding to microsomal protein, collagen, ligandin and nucleic acids in vitro. The amino acid residue which reacts with zomepirac glucuronide and the mechanism for the binding will be examined. Factors which influence the exposure to the reactive glucuronides in vivo will be examined by testing the effects of reduced excretion (renal failure and cholestasis); increased rate of glucuronide formation (phenobarbital); and reduced hydrolysis by esterase inhibition (PMSF, pyridostigmine), on the extent of irreversible binding in vivo in animals. Finally, studies will be conducted with young, healthy human volunteers and elderly patients to determine if covalent binding occurs in vivo for acidic drugs presently in clinical use, and whether an increase in the extent of covalent binding is correlated with exposure to the acyl glucuronide.